We report on a preliminary experimental study of an interferometer built with two 500 meters long arms made of polarization maintaining optical fibers. The control of the field polarization state along the single-mode fiber arms enables to measure fringe contrast up to 93% with a laser source emitting a 1290nm carrier wavelength. Tests achieved with broadband spectrum exhibit dispersion differential effect resulting from fiber inhomogeneities. Partial compensation of this effect is achieved introducing additional fiber pieces on one arm.
©2001 Optical Society of America
In a synthesized aperture implemented for stellar interferometry applications, the use of optical fiber to carry the light beams from telescopes to the mixing station is a good alternative to the conventional device consisting of vacuum or air tubes and mirror trains [1,2]. Main advantages of fiber use come from their high transmission of light flux, the flexibility of these waveguides and spatial filtering which reduces the effects of the atmospheric turbulence and others instrument optical aberrations [3,4].
Nevertheless, to preserve the optical field coherence all along the fiber, various effects must be compensated or controlled. First, the dispersive properties of fibers lead to differential effects reducing the fringe visibility contrast for interferometric applications [5,6]. Second, the high sensitivity of fibers to thermal and mechanical perturbations induces difficulties to control the polarization state [7,8] but also compels to control the fiber optical length with a metrological device [9,10]. Experimentally, for an interferometric device including three 25-m-long highly birefringent (HB) fiber arms, full control of optical path stability and correction of dispersion and polarization differential effects have been achieved. It led to the observation of well-contrasted fringes with a randomly polarized source radiating a broadband spectrum (650–850nm) . On the other hand, the various techniques developed at the IRCOM Institute led to the implementation of an all guided interferometer using fiber delay line to synchronize the fields and fiber couplers for the beam recombination .
At the present time, several projects plan to connect elementary telescopes of a longbaseline interferometer over hundreds meters such as OHANA project [13–15]. Then, we purpose to experimentally investigate the potentiality of silica fiber to propagate coherently light beams over kilometric spans.
2. General Description of the Experiment
The experimental set up is shown in the figure 1 and consists of a Mach-Zehnder interferometer. Each interferometric arm is implemented with a 500 meter long HB1250 polarization maintaining optical fiber with a λc=1090nm cut-off wavelength and a 1mm beat-length. With a very similar fiber, the dispersion at 1270 nm is in the range of -10 ps/nm.km. The stress-applying part of this fiber is ‘Bow-Tie’ type with an extinction ration in the range of 20dB. These two fiber arms are came from the same fiber roll in order to minimize all differential effects. The manufacturer matches the two sections of fiber in geometrical length with twenty centimeters accuracy.
Our interferometer can be alternately fed by two light sources. The first one is a laser diode emitting a 1290nm mean wavelength. The coherence length of this source can be modified by adjusting the driving current. We obtain either a narrow spectrum with 0.1nm spectral bandwidth or, under the laser threshold, a broadband emission with a 30nm spectral bandwidth. The second source consists of a LED with a 140nm spectral bandwidth around 1270nm. These wavelengths have been intentionally selected with a view to taking advantage of the low chromatic dispersion effects of silica fibers in these spectral domains. For other wavelength, the material dispersion increases. It could involve difficulties to locate the zero group delay between the two fiber arms because the fast degradation of fringe contrast if the dispersion characteristics of the two fiber sections are not accurately matched. These various sources are not spatially coherent. We insert a standard single-mode fiber between the source and the collimator focal plane to avoid reducing the measured fringe visibility by spatial coherence effects. The emerging beam is linearly polarized with a polarizer. After passing through the collimator, the optical field is partly collected by two microscope objectives acting as telescopes and located in the same wave front. These apertures feed the fiber inputs taking care to align the linearly polarized optical field with the fast axis of the two input sections of HB fibers. Each fiber output is placed on a beam collimating assembly including a microscope objective.
One of the interferometric fiber arms has a 4-m fiber length wound over a 60-mm-diameter cylindrical piezoelectric transducer (PZT). This modulator generates an optical path modulation used to temporally scan the fringes. The full stroke of this fiber stretcher is in the range of 100µm. In addition, an optical fiber delay line [6,12] has been implemented with a 12cm total stroke to control the differential group delay in the interferometer with an accurate resolution. This fiber delay line consists of seven meters of HB fiber wound and glued on a rubber rim. A mechanical process drives the rubber rim radial expansion so that the fiber is uniformly stretched. The total geometrical length of the first fiber arm of the interferometer is close to 500m. The second interferometric arm includes 500m of HB fiber and an air delay based on the output beam collimating assemblies motion by steps of 10cm. A beam splitter achieves the recombination of the linearly polarized output beams. This interferometric mixing is launched in a standard single-mode fiber to perform a spatial filtering that cancels the influence of beams overlapping quality on the fringe visibility contrast. The emerging beam is focused on an InGaAs monopixel detector.
3. Experimental Results
In a first experimental step, we have characterized the cross-coupling between the two principal polarization modes for each interferometric arm. The following relation defines the extinction ratio in decibel:
Where Ix is the intensity of excited mode and Iy the parasitic intensity coupled to the second polarization mode. For the two interferometric arms, this ratio was better than 15db throughout the experiment. It will lead to a 5% theoretical loss of fringe contrast.
In the second experimental step, we have looked at the fringe position by canceling the group delay between the two 500m interferometric arms. For this purpose, the laser diode is used under the threshold leading a short coherence length. We have scanned the fringe pattern by using the fiber PZT modulator. The equalization of the group delay was achieved by actuating the fiber delay line placed in the first interferometric arm for various positions of the output collimating assembly of the second fiber arm. The fringes have been found with a 20 cm air delay. If we consider homogeneous fiber, the corresponding difference in geometrical length between the two interferometric arms would be 13cm. The flux on each recombined beams are alternately measured to take the photometric effects on the fringe visibility measurements into account. When the laser operates in the narrow spectral bandwidth mode, the measured contrast is equal to 92%. With this source, the differential chromatic dispersion effects are not significant. Consequently, this high level of visibility contrast demonstrates the good control of the polarization status of the propagating waves. The fringe contrasts measured with large spectrum sources i.e. 30nm and 120nm are close to 24% and 11% respectively. These low contrasts observed with a broadband spectrum source can be justified by a strong chromatic dispersion effect between the two fiber arms. The inhomogeneity of the waveguide structure can mainly explain this differential effect of dispersion over this spectral domain where the material dispersion is near zero .
The third experimental step was based on the visibility contrast improvement with a broadband spectrum source by adding various HB single-mode fiber pieces in one of the arms. To search the least dispersive configuration, we alternately plug a fiber sample on each interferometric arm and we measure the fringe contrast evolution. We equalize the two optical paths of the interferometer by adjusting the air delay in order to roughly cancel the equivalent air path generated by the additional fiber sample. Then we accurately retrieve the fringe packet around the zero group delay by continuously actuating the optical fiber delay line. Figure 2 reports on the different visibility measurements got for various spectral bandwidths as a function of the equivalent air path associated to additional fiber samples. The geometrical lengths of these additional fiber samples varied between 0.5m and 7.2m. The weak dispersion of the silica fiber material around 1300nm justifies the use of large geometrical length of additional sample in order to compensate the dispersion between the two interferometric fiber arms.
With the narrow spectrum source, the fringe visibility is close to 93% with a 1.3% standard deviation whatever the additional fiber sample. These high levels of contrast prove once again the good control of the polarization state all along the propagation in the interferometer. The visibility measurement deviation is mainly due to the large fluctuations of the intensity that is not monitored in real time. The maximum of contrast observed with the two broadband sources are 69% and 42% for a 30nm and 140nm spectral bandwidth respectively. These contrasts are obtained with a 5.2m long additional fiber sample. Owing to the fact that the field polarization state are mastered, the degradation of the visibility contrast according to the spectral bandwidth can be explain by the variable dispersion between the two fiber arms resulting from inhomogeneities.
The experimental results show that the differential dispersion effects cannot be cancelled only by adjusting the difference in optical length between the two fibers. Consequently an additional experimental study is necessary to achieve a spectral analysis to determine the differential chromatic dispersion evolution.
We have experimentally demonstrated the possibility to obtain well-contrasted fringes in a kilometric interferometer. The use of polarization maintaining fiber enables to control the polarization state leading to high contrast measurement with a laser source. Nevertheless, the dispersion inhomogeneity of the fibers induces fringe contrast reduction for a source with a broadband spectrum. A partial compensation allowed us to minimize this effect, the contrast reaching 42% for a 140nm spectral bandwidth around 1270nm.
To continue this study, we plan an experimental investigation to characterize the differential dispersion evolution according to the various additional fiber sections. A spectral analysis of the interferometric mixing using the channeled spectrum method  will allow to accurately measure the differential effect of chromatic dispersion i.e. second and third order term of the phase shift versus frequency. Only this analysis will allow modelling the dispersion characteristics and the corresponding evolution of fringe contrast. Moreover, we schedule a study on the metrological system devoted to control the optical path difference in the frame of a long-baseline interferometer with kilometric arms. This interferometric device can be a good solution to perform the connection of several elementary telescopes such as in the OHANA project or in the frame of large spatial interferometer.
References and links
1. J. Davis, A. Mendez, J. W. O’Byrne, E. B. Seneta, W. J. Tango, and P. G. Tuthill, “Sydney University Stellar Interferometer Program,” In Interferometry in Optical Astronomy, P. J. Léna and A. Quirrenbach, eds., Proc SPIE4006, 434–443 (2000).
2. C. A. Haniff, J. E. Baldwin, R. C. Boysen, A. V. George, D. F. Buscher, C. D. Mackay, D. Pearson, J. Rogers, P. J. Warner, D. M. AA. Wilson, and J. S. Young, “COAST - the current status,” In Interferometry in Optical Astronomy, P. J. Léna and A. Quirrenbach, eds., Proc SPIE4006, 627–633 (2000).
3. S. Shaklan and F. Roddier, “Single-mode fiber optics in a long-baseline interferometer”, Appl. Opt 26, 2334–2338 (1988). [CrossRef]
4. V. Coudé du Foresto and S. Rigway, “A stellar interferometer using single-mode infrared fibers,” in High-resolution imaging by inteferometry II, J. Beckers et and F. Merkle, eds., ESO, Garching, Germany, 731–740 (1991).
5. S. D. Dyer and D. A. Christensen, “Dispersion effects in fiber optic interferometry,” Opt. Eng. 36, 2440–2447 (1997). [CrossRef]
6. L. M. Simohamed and F. Reynaud, “Characterization of the dispersion evolution versus stretching in a large stroke optical fiber delay line,” Opt. Commun. 159, 118–128 (1999). [CrossRef]
7. S.C. Rashleigh, “Origins and control of polarization effects in single-mode fibers,” Jour. of Lightw. Tech. LT-1, 312–330 (1983). [CrossRef]
8. L. Delage and F. Reynaud, “Analysis and control of polarization effects on phase closure and image acquisition in a fiber-linked three-telescope stellar interferometer,” J. Opt. A 2, 1–7 (2000). [CrossRef]
10. L. Delage, F. Reynaud, and A. Lannes, “Laboratory imaging stellar interferometer with fiber links,” Appl. Opt. 39, 6406–6420 (2000). [CrossRef]
11. F. Reynaud and H. Lagorceix, “Stabilization and control of a fiber array for the coherent transport of beams in a stellar interferometer,” in Proceedings of the Conference Astrofib’96 on Integrated Optics for Astronomical Interferometry, P. Kern and F. Malbet, Eds. (Bastianelli-Guirimaud, Grenoble, 1996), 249–257 (1997).
12. G. Huss, M. L. Simohamed, and F. Reynaud, “An all guided two-beam stellar interferometer: preliminary experiment,” Opt. Commun. 182, 71–82 (2000). [CrossRef]
13. J-M. Mariotti, V. Coudé du Foresto, G. Perrin, Peiqian Zhao, and P. Léna, “Interferometric connection of large ground-based telescopes,” Astronomy& Astrophysics, Supplement series , 116, 381–393 (1996). [CrossRef]
14. G. Perrin, O. Lay, P. Léna, and V. Coudé du Foresto, “A fibered large interferometer on top of Mauna Kea : OHANA, the Optical Hawaiian Array for Nano-radian Astronomy,” In Interferometry in Optical Astronomy, P. J. Léna and A. Quirrenbach, eds., Proc SPIE4006, 708–714 (2000).
15. K. Sato, J. Nishikawa, M. Yoshizawa, T. Fukushima, Y. Torii, K. Matsuda, K. Kubo, H. Iwashita, S. Suzuki, and D. Saint-Jacques, “Experiments of the fiber-connected interferometer for MIRA project,” In Interferometry in Optical Astronomy, P. J. Léna and A. Quirrenbach, eds., Proc SPIE4006, 1102–1106 (2000).
16. H.T. Shang, “Chromatic dispersion measurement by white-light interferometry on metre-length single-mode optical fibres,” Electronic Lett. 17, 603–605 (1981). [CrossRef]